Face Identification Through Facial Skeletal Features

Abstract

AbstractAutomatic face identification technology has received lots of attention in the last few decades in the field of image analysis and computer vision. Automatic face identification is a challenging problem in image analysis. Psychologists and neurologists analyze facial images to discover human behavior and personality. Computer scientists deal with computation like feature extraction, similarity measurement, classification, etc., on facial images. Statistical shape analysis is one of the crucial areas in computer vision. Various face features are collected and then used for identification purposes or to find the asymmetry between the two parts of the face. The proposed study presents an algorithm for the face identification model using the distance between various skeletal facial features like mouth, eyes, and nose. For finding the facial features, the Viola–Jones face detection algorithm is used. K-nearest neighbor (KNN), Naive Bayes, and Bootstrap aggregation classifiers are used for the purpose of identification. The proposed model has been tested on the FEI, Faces94, Faces95, Grimace, BioID, and CVL datasets. The proposed algorithm is implemented in MATLAB and gives 94.33% accuracy.KeywordsComputer visionViola-JonesFace identificationNearest neighborNaive BayesBAG